Automatic recognition of vigilance state by using a wavelet-based artificial neural network

In this study, 5-s long sequences of full-spectrum electroencephalogram (EEG) recordings were used for classifying alert versus drowsy states in an arbitrary subject. EEG signals were obtained from 30 healthy subjects and the results were classified using a wavelet-based neural network. The wavelet-based neural network model, employing the multilayer perceptron (MLP), was used for the classification of EEG signals. A multilayer perceptron neural network (MLPNN) trained with the Levenberg–Marquardt algorithm was used to discriminate the alertness level of the subject. In order to determine the MLPNN inputs, spectral analysis of EEG signals was performed using the discrete wavelet transform (DWT) technique. The MLPNN was trained, cross-validated, and tested with training, cross-validation, and testing sets, respectively. The correct classification rate was 93.3% alert, 96.6% drowsy, and 90% sleep. The classification results showed that the MLPNN trained with the Levenberg–Marquardt algorithm was effective for discriminating the vigilance state of the subject.     

Detecting and connecting disjoint sub-networks in wireless sensor and actor networks

Wireless sensor and actor networks (WSANs) can be considered as a combination of a sensor network and an actor network in which powerful and mobile actor nodes can perform application specific actions based on the received data from the sensors. As most of these actions are performed collaboratively among the actors, inter-actor connectivity is one of the desirable features of WSANs. In this paper, we propose a novel distributed algorithm for establishing a connected inter-actor network topology. Considering initially disjoint sets of actors, our algorithm first initiates a search process by using the underlying sensor network in order to detect the possible sub-networks of actors in the region. After these sub-networks are detected, our algorithm pursues a coordinated actor movement in order to connect the sub-networks and thus achieve inter-actor connectivity for all the actors. This coordinated movement approach exploits the minimum connected dominating set of each sub-network when picking the appropriate actor to move so that the connectivity of each sub-network is not violated. In addition, the approach strives to minimize the total travel distance of actors and the messaging cost on both sensors and actors in order to extend the lifetime of WSAN. We analytically study the performance of our algorithm. Extensive simulation experiments validate the analytical results and confirm the effectiveness of our approach.     

Microcystin-LR toxicity on dominant copepods Eurytemora affinis and Pseudodiaptomus forbesi of the upper San Francisco Estuary

This study investigates the toxicity and post-exposure effects of dissolved microcystin (MC-LR) on the dominant copepods of the upper San Francisco Estuary (SFE), where blooms of the toxic cyanobacteria Microcystis aeruginosa coincide with record low levels in the abundance of pelagic organisms including phytoplankton, zooplankton, and fish. The potential negative impact of Microcystis on the copepods Eurytemora affinis and Pseudodiaptomus forbesi has raised concern for further depletion of high quality fish food. Response of copepods to MC-LR (MC) was determined using a 48-h standard static renewal method for acute toxicity testing. Following exposure, a life table test was performed to quantify any post-exposure impacts on survival and reproduction. The 48-h LC-50 and LC-10 values for MC were 1.55 and 0.14 mg/L for E. affinis; and 0.52 and 0.21 mg/L for P. forbesi. Copepod populations recovered once dissolved MC was removed and cultures returned to optimal conditions, suggesting no post-exposure effects of MC on copepod populations. Dissolved microcystin above 0.14 mg/L proved likely to have chronic effects on the survival of copepods in the SFE. Since such high concentrations are unlikely, toxicity from dissolved microcystin is not a direct threat to zooplankton of the SFE, and other mechanisms such as dietary exposure to Microcystis constitute a more severe risk.     

Effect of Messaging Model on the Reliable Data Transfer Latency in a Fog System

Journal of Network and Systems Management - IoT applications have become a pillar for enhancing the quality of life. However, the increasing amount of data generated by IoT devices places pressure...     

Deep long short-term memory networks-based automatic recognition of six different digital modulation types under varying noise conditions

Neural Computing and Applications - In this paper, a new method based on deep learning has been proposed in order to recognize noise-digital modulation signals at varying noise levels...     

Strong Damping‐Like Spin‐Orbit Torque and Tunable Dzyaloshinskii–Moriya Interaction Generated by Low‐Resistivity Pd1−xPtx Alloys

Despite their great promise for providing a pathway for very efficient and fast manipulation of magnetization, spin‐orbit torque (SOT) operations are currently energy inefficient due to a low damping‐like SOT efficiency per unit current bias, and/or the very high resistivity of the spin Hall materials. This work reports an advantageous spin Hall material, Pd_1?xPt_x, which combines a low resistivity with a giant spin Hall effect as evidenced with three independent SOT ferromagnetic detectors. The optimal Pd_0.25Pt_0.75 alloy has a giant internal spin Hall ratio of >0.60 (damping‐like SOT efficiency of ≈0.26 for all three ferromagnets) and a low resistivity of ≈57.5 µΩ cm at a 4 nm thickness. Moreover, it is found that the Dzyaloshinskii–Moriya interaction (DMI), the key ingredient for the manipulation of chiral spin arrangements (e.g., magnetic skyrmions and chiral domain walls), is considerably strong at the Pd_1?xPt_x/Fe_0.6Co_0.2B_0.2 interface when compared to that at Ta/Fe_0.6Co_0.2B_0.2 or W/Fe_0.6Co_0.2B_0.2 interfaces and can be tuned by a factor of 5 through control of the interfacial spin‐orbital coupling via the heavy metal composition. This work establishes a very effective spin current generator that combines a notably high energy efficiency with a very strong and tunable DMI for advanced chiral spintronics and spin torque applications.     

A separable augmented Lagrangian algorithm for optimal structural design

Structural and Multidisciplinary Optimization - We propose an iterative separable augmented Lagrangian algorithm (SALA) for optimal structural design, with SALA being a subset of the alternating...     

Equivalent circuit for capacitive micromachined ultrasonic transducers to predict anti-resonances

Equivalent circuit models have been long used to evaluate the dynamics of the capacitive micromachined ultrasonic transducer (CMUT). An important parameter in the characterization of a CMUT is the anti-resonance frequency, which limits the immersion bandwidth. However, there is no equivalent circuit model that can accurately determine the anti-resonance frequency of a membrane. In this work, we present an improved lumped element parametric model for immersed CMUT. We demonstrate that the proposed equivalent circuit model accurately predicts anti-resonance and higher order mode frequencies, in addition to that of the fundamental mode. The proposed circuit model is in good agreement with device characteristics calculated using the finite element method and experimentally measured data.

     

RETRACTED ARTICLE: Tolerance rough set firefly-based quick reduct

In medical information system, there are a lot of features and the relationship among elements is solid. In this way, feature selection of medical datasets gets awesome worry as of late. In this article, tolerance rough set firefly-based quick reduct, is developed and connected to issue of differential finding of diseases. The hybrid intelligent framework intends to exploit the advantages of the fundamental models and, in the meantime, direct their restrictions. Feature selection is procedure for distinguishing ideal feature subset of the original features. A definitive point of feature selection is to build the precision, computational proficiency and adaptability of expectation strategy in machine learning, design acknowledgment and information mining applications. Along these lines, the learning framework gets a brief structure without lessening the prescient precision by utilizing just the chose remarkable features. In this research, a hybridization of two procedures, tolerance rough set and as of late created meta-heuristic enhancement calculation, the firefly algorithm is utilized to choose the conspicuous features of medicinal information to have the capacity to characterize and analyze real sicknesses. The exploratory results exhibited that the proficiency of the proposed system outflanks the current supervised feature selection techniques.